Question 1

Can the AMD Ryzen AI Max (Strix Halo) run qwen2.5vl 32B?

Accepted Answer

qwen2.5vl 32B needs about 23GB working set (weights 22GB + KV cache 1GB at 4k context). The AMD Ryzen AI Max (Strix Halo) offers ~30GB usable — verdict: full. 23.0 GB needed of 30 GB usable — headroom for context

Question 2

How fast is qwen2.5vl 32B on the AMD Ryzen AI Max (Strix Halo)?

Accepted Answer

~7–9.9 tok/s on the AMD Ryzen AI Max (Strix Halo). computed roofline: 256 GB/s × 0.60–0.85 efficiency window / 22.0GB (Q4_K_M) — bands, never points.

Question 3

What should I run qwen2.5vl 32B on instead?

Accepted Answer

Mac mini M5 Pro (64GB) runs it fully ($1669).

Can the AMD Ryzen AI Max (Strix Halo) run qwen2.5vl 32B?

Run qwen2.5vl 32B on instead